Simon P. Couch comments

Results 165 comments of


                                            Simon P. Couch

`collect_metrics` ignores case weights in calculating `concordance_survival`

Going to go ahead and close. Feel free to ping here or in a separate issue if you think this default ought to be changed, as we're very much interested...

Regression running tune_bayes on mars with tune 1.1.2

Thanks for the issue! This is a strange one. With tune 1.1.1, I see: ``` r two_bayer_res %>% collect_metrics() #> # A tibble: 8 × 8 #> num_terms .metric .estimator...

Regression running tune_bayes on mars with tune 1.1.2

Ah, yes. With 1.1.2, if you set `control = control_bayes(TRUE, TRUE)`, then you'll see the errors go away, as before. The underlying issue as that the GP can't predict any...

tune_bayes does not finalize by itself

Thanks for the issue! I don't have access to your `bt_wf` and `comb4_set` objects. Can you please provide a minimal [reprex](https://github.com/tidyverse/reprex#what-is-a-reprex) (reproducible example)? A reprex will help me troubleshoot and...

Hard-to-understand error message with CV and missing data

Related to tidymodels/parsnip#812, though I agree that they're unique and ought to live where they do.🌻 Different engines fail to different degrees here, so fixes may need to live in...

Make the `.config` column in `collect_metrics()` output more informative

Agreed, thank you for the issue! I feel similarly that it should be straightforward to make those names more informative, though tune, finetune, workflowsets, and stacks all have machinery to...

first pass at postprocessing proof-of-concept

On 3.6.3, seeing: ``` Error: ! error in pak subprocess Caused by error: ! Could not solve package dependencies: * deps::.: Can't install dependency tidymodels/container#1 * tidymodels/container#1: Can't install dependency...

first pass at postprocessing proof-of-concept

I believe the dependency conflicts in GHA installs right now are due to the fact that `tidymodels/recipes` and `tidymodels/parsnip` have a `tidymodels/hardhat` (as in `main`) `Remotes` ref and this one...

first pass at postprocessing proof-of-concept

Turns out this PR doesn't need further changes to support tidymodels/container#12. With that PR and this PR as-is: ``` r library(tidymodels) library(container) library(probably) #> #> Attaching package: 'probably' #> The...

first pass at postprocessing proof-of-concept

[ec0effa](https://github.com/tidymodels/workflows/pull/225/commits/ec0effa0a8e9d46faefb3616b65bae3599e0620a) surfaces an important point; removing/updating a postprocessor from an otherwise trained workflow need not remove the preprocessor and model fits, as they won't be affected by the removal of...